Daily Address Scraper Needed

I need a reliable script that crawls every public page on http://oeffnungszeitenbuch.de/ and extracts the full Address field from each business listing. Nothing else—just the raw street address exactly as it appears on the site.

The scraper must:
• Traverse the entire site automatically each day, capturing new or updated listings without duplicates.
• Export the complete set of addresses to a single Excel (.xlsx) file and overwrite or version it daily.
• Run unattended on a Linux VPS (Python + Scrapy, Playwright, or similar is fine as long as it is head-error-tolerant and respects polite crawling speeds).
• Log any failed pages so I can spot-check issues quickly.

Deliverables:
1. Well-documented source code.
2. A brief README with installation steps and the command (or cron) I will run to trigger the daily job.
3. One sample Excel file produced by your script showing a handful of extracted addresses.

I will consider the job complete once the script runs for 24 hours on my server, generates the Excel file without missing sections of the site, and the log shows zero critical errors.